Children with cardiac arrest (CA) have mortality rates of 50-90%, largely due to neurological failure as part of the post-resuscitation syndrome. There is a critical gap of knowledge and tools to accurately classify outcome after pediatric CA. Physical examination and laboratory testing inadequately assess the severity of neurologic injury and outcome. Hazards of misclassification include risking adverse effects from ineffective therapies and non-treatment of ostensibly well patients who later are found to have neurologic deficits. Early and accurate identification of the eventual severity of neurologic injury would allw for timely neuroprotective interventions and/or more targeted testing of new therapies in specific risk populations. Our long term objective is to improve the neurological outcome of children surviving CA. Here we propose to model and validate serum and imaging biomarkers of brain injury with empirical support, and to assess their accuracy together with clinical variables in classifying outcome after pediatric CA. We seek to capitalize on robust preliminary data from an NIH-funded single center RCT in pediatric CA and our track record in biomarker research in pediatric brain injury. Our central hypothesis is that serum and imaging biomarkers of brain injury, together with clinical variables, will critically aid in the early classification of favorale outcome after pediatric CA (Vineland Adaptive Behavior Scales score > 70) 1 year after pediatric CA in a multicenter prospective study (8 centers and 248 subjects). Strong preliminary data supports this hypothesis, and biomarkers will be tested for outcome classification accuracy in the following 3 specific aims: Aim 1) Serum biomarkers of neuronal (neuron specific enolase and ubiquitin carboxy-terminal hydrolase-L1) and glial injury (S100b and glial fibrillary acidic protein); Aim 2) Regional (occipital-parietal cortex, basal ganglia, and thalamus) brain MRI (T1/T2 and diffusion-weighted imaging) and MR spectroscopy biomarkers of neuronal injury (N-acetyl-aspartate) and energy failure (lactate); and Aim 3 will model the combination of strong serum and imaging biomarkers of brain injury with clinical variables. We will assess serum biomarkers of brain mitochondrial injury with potential for novel therapeutic targets (cardiolipin and oxidized cardiolipin) in an exploratory aim. This proposed research is innovative, because we will prospectively develop and optimize a combined panel of serum and imaging biomarkers with clinical variables to accurately classify outcome after pediatric CA. These proposed aims leverage recent pilot successes and should generate accurate and reliable models of biomarkers that markedly improve post-resuscitation clinical care in children after CA. Furthermore, these results are expected to have a positive impact in advancing neurocritical care for these children, with forthcoming development of a serum biomarker point of care test and biomarker panels that will accurately classify risk of unfavorable outcome for clinicians and researchers needing to stratify by severity of injury, to monitor response to therapy, and ultimately to assist in their rehabilitation and recovery.